Merge tag 'net-next-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-nextgrafted

Pull networking updates from Jakub Kicinski:
 "Core:

   - Add dedicated kmem_cache for typical/small skb->head, avoid having
     to access struct page at kfree time, and improve memory use.

   - Introduce sysctl to set default RPS configuration for new netdevs.

   - Define Netlink protocol specification format which can be used to
     describe messages used by each family and auto-generate parsers.
     Add tools for generating kernel data structures and uAPI headers.

   - Expose all net/core sysctls inside netns.

   - Remove 4s sleep in netpoll if carrier is instantly detected on
     boot.

   - Add configurable limit of MDB entries per port, and port-vlan.

   - Continue populating drop reasons throughout the stack.

   - Retire a handful of legacy Qdiscs and classifiers.

  Protocols:

   - Support IPv4 big TCP (TSO frames larger than 64kB).

   - Add IP_LOCAL_PORT_RANGE socket option, to control local port range
     on socket by socket basis.

   - Track and report in procfs number of MPTCP sockets used.

   - Support mixing IPv4 and IPv6 flows in the in-kernel MPTCP path
     manager.

   - IPv6: don't check net.ipv6.route.max_size and rely on garbage
     collection to free memory (similarly to IPv4).

   - Support Penultimate Segment Pop (PSP) flavor in SRv6 (RFC8986).

   - ICMP: add per-rate limit counters.

   - Add support for user scanning requests in ieee802154.

   - Remove static WEP support.

   - Support minimal Wi-Fi 7 Extremely High Throughput (EHT) rate
     reporting.

   - WiFi 7 EHT channel puncturing support (client & AP).

  BPF:

   - Add a rbtree data structure following the "next-gen data structure"
     precedent set by recently added linked list, that is, by using
     kfunc + kptr instead of adding a new BPF map type.

   - Expose XDP hints via kfuncs with initial support for RX hash and
     timestamp metadata.

   - Add BPF_F_NO_TUNNEL_KEY extension to bpf_skb_set_tunnel_key to
     better support decap on GRE tunnel devices not operating in collect
     metadata.

   - Improve x86 JIT's codegen for PROBE_MEM runtime error checks.

   - Remove the need for trace_printk_lock for bpf_trace_printk and
     bpf_trace_vprintk helpers.

   - Extend libbpf's bpf_tracing.h support for tracing arguments of
     kprobes/uprobes and syscall as a special case.

   - Significantly reduce the search time for module symbols by
     livepatch and BPF.

   - Enable cpumasks to be used as kptrs, which is useful for tracing
     programs tracking which tasks end up running on which CPUs in
     different time intervals.

   - Add support for BPF trampoline on s390x and riscv64.

   - Add capability to export the XDP features supported by the NIC.

   - Add __bpf_kfunc tag for marking kernel functions as kfuncs.

   - Add cgroup.memory=nobpf kernel parameter option to disable BPF
     memory accounting for container environments.

  Netfilter:

   - Remove the CLUSTERIP target. It has been marked as obsolete for
     years, and we still have WARN splats wrt races of the out-of-band
     /proc interface installed by this target.

   - Add 'destroy' commands to nf_tables. They are identical to the
     existing 'delete' commands, but do not return an error if the
     referenced object (set, chain, rule...) did not exist.

  Driver API:

   - Improve cpumask_local_spread() locality to help NICs set the right
     IRQ affinity on AMD platforms.

   - Separate C22 and C45 MDIO bus transactions more clearly.

   - Introduce new DCB table to control DSCP rewrite on egress.

   - Support configuration of Physical Layer Collision Avoidance (PLCA)
     Reconciliation Sublayer (RS) (802.3cg-2019). Modern version of
     shared medium Ethernet.

   - Support for MAC Merge layer (IEEE 802.3-2018 clause 99). Allowing
     preemption of low priority frames by high priority frames.

   - Add support for controlling MACSec offload using netlink SET.

   - Rework devlink instance refcounts to allow registration and
     de-registration under the instance lock. Split the code into
     multiple files, drop some of the unnecessarily granular locks and
     factor out common parts of netlink operation handling.

   - Add TX frame aggregation parameters (for USB drivers).

   - Add a new attr TCA_EXT_WARN_MSG to report TC (offload) warning
     messages with notifications for debug.

   - Allow offloading of UDP NEW connections via act_ct.

   - Add support for per action HW stats in TC.

   - Support hardware miss to TC action (continue processing in SW from
     a specific point in the action chain).

   - Warn if old Wireless Extension user space interface is used with
     modern cfg80211/mac80211 drivers. Do not support Wireless
     Extensions for Wi-Fi 7 devices at all. Everyone should switch to
     using nl80211 interface instead.

   - Improve the CAN bit timing configuration. Use extack to return
     error messages directly to user space, update the SJW handling,
     including the definition of a new default value that will benefit
     CAN-FD controllers, by increasing their oscillator tolerance.

  New hardware / drivers:

   - Ethernet:
      - nVidia BlueField-3 support (control traffic driver)
      - Ethernet support for imx93 SoCs
      - Motorcomm yt8531 gigabit Ethernet PHY
      - onsemi NCN26000 10BASE-T1S PHY (with support for PLCA)
      - Microchip LAN8841 PHY (incl. cable diagnostics and PTP)
      - Amlogic gxl MDIO mux

   - WiFi:
      - RealTek RTL8188EU (rtl8xxxu)
      - Qualcomm Wi-Fi 7 devices (ath12k)

   - CAN:
      - Renesas R-Car V4H

  Drivers:

   - Bluetooth:
      - Set Per Platform Antenna Gain (PPAG) for Intel controllers.

   - Ethernet NICs:
      - Intel (1G, igc):
         - support TSN / Qbv / packet scheduling features of i226 model
      - Intel (100G, ice):
         - use GNSS subsystem instead of TTY
         - multi-buffer XDP support
         - extend support for GPIO pins to E823 devices
      - nVidia/Mellanox:
         - update the shared buffer configuration on PFC commands
         - implement PTP adjphase function for HW offset control
         - TC support for Geneve and GRE with VF tunnel offload
         - more efficient crypto key management method
         - multi-port eswitch support
      - Netronome/Corigine:
         - add DCB IEEE support
         - support IPsec offloading for NFP3800
      - Freescale/NXP (enetc):
         - support XDP_REDIRECT for XDP non-linear buffers
         - improve reconfig, avoid link flap and waiting for idle
         - support MAC Merge layer
      - Other NICs:
         - sfc/ef100: add basic devlink support for ef100
         - ionic: rx_push mode operation (writing descriptors via MMIO)
         - bnxt: use the auxiliary bus abstraction for RDMA
         - r8169: disable ASPM and reset bus in case of tx timeout
         - cpsw: support QSGMII mode for J721e CPSW9G
         - cpts: support pulse-per-second output
         - ngbe: add an mdio bus driver
         - usbnet: optimize usbnet_bh() by avoiding unnecessary queuing
         - r8152: handle devices with FW with NCM support
         - amd-xgbe: support 10Mbps, 2.5GbE speeds and rx-adaptation
         - virtio-net: support multi buffer XDP
         - virtio/vsock: replace virtio_vsock_pkt with sk_buff
         - tsnep: XDP support

   - Ethernet high-speed switches:
      - nVidia/Mellanox (mlxsw):
         - add support for latency TLV (in FW control messages)
      - Microchip (sparx5):
         - separate explicit and implicit traffic forwarding rules, make
           the implicit rules always active
         - add support for egress DSCP rewrite
         - IS0 VCAP support (Ingress Classification)
         - IS2 VCAP filters (protos, L3 addrs, L4 ports, flags, ToS
           etc.)
         - ES2 VCAP support (Egress Access Control)
         - support for Per-Stream Filtering and Policing (802.1Q,
           8.6.5.1)

   - Ethernet embedded switches:
      - Marvell (mv88e6xxx):
         - add MAB (port auth) offload support
         - enable PTP receive for mv88e6390
      - NXP (ocelot):
         - support MAC Merge layer
         - support for the the vsc7512 internal copper phys
      - Microchip:
         - lan9303: convert to PHYLINK
         - lan966x: support TC flower filter statistics
         - lan937x: PTP support for KSZ9563/KSZ8563 and LAN937x
         - lan937x: support Credit Based Shaper configuration
         - ksz9477: support Energy Efficient Ethernet
      - other:
         - qca8k: convert to regmap read/write API, use bulk operations
         - rswitch: Improve TX timestamp accuracy

   - Intel WiFi (iwlwifi):
      - EHT (Wi-Fi 7) rate reporting
      - STEP equalizer support: transfer some STEP (connection to radio
        on platforms with integrated wifi) related parameters from the
        BIOS to the firmware.

   - Qualcomm 802.11ax WiFi (ath11k):
      - IPQ5018 support
      - Fine Timing Measurement (FTM) responder role support
      - channel 177 support

   - MediaTek WiFi (mt76):
      - per-PHY LED support
      - mt7996: EHT (Wi-Fi 7) support
      - Wireless Ethernet Dispatch (WED) reset support
      - switch to using page pool allocator

   - RealTek WiFi (rtw89):
      - support new version of Bluetooth co-existance

   - Mobile:
      - rmnet: support TX aggregation"

* tag 'net-next-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next: (1872 commits)
  page_pool: add a comment explaining the fragment counter usage
  net: ethtool: fix __ethtool_dev_mm_supported() implementation
  ethtool: pse-pd: Fix double word in comments
  xsk: add linux/vmalloc.h to xsk.c
  sefltests: netdevsim: wait for devlink instance after netns removal
  selftest: fib_tests: Always cleanup before exit
  net/mlx5e: Align IPsec ASO result memory to be as required by hardware
  net/mlx5e: TC, Set CT miss to the specific ct action instance
  net/mlx5e: Rename CHAIN_TO_REG to MAPPED_OBJ_TO_REG
  net/mlx5: Refactor tc miss handling to a single function
  net/mlx5: Kconfig: Make tc offload depend on tc skb extension
  net/sched: flower: Support hardware miss to tc action
  net/sched: flower: Move filter handle initialization earlier
  net/sched: cls_api: Support hardware miss to tc action
  net/sched: Rename user cookie and act cookie
  sfc: fix builds without CONFIG_RTC_LIB
  sfc: clean up some inconsistent indentings
  net/mlx4_en: Introduce flexible array to silence overflow warning
  net: lan966x: Fix possible deadlock inside PTP
  net/ulp: Remove redundant ->clone() test in inet_clone_ulp().
  ...

1 files changed, 517 insertions, 0 deletions

diff --git a/lib/mpi/mpih-div.c b/lib/mpi/mpih-div.c
new file mode 100644
index 000000000..be70ee2e4
--- /dev/null
+++ b/lib/mpi/mpih-div.c
@@ -0,0 +1,517 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/* mpihelp-div.c  -  MPI helper functions
+ *	Copyright (C) 1994, 1996 Free Software Foundation, Inc.
+ *	Copyright (C) 1998, 1999 Free Software Foundation, Inc.
+ *
+ * This file is part of GnuPG.
+ *
+ * Note: This code is heavily based on the GNU MP Library.
+ *	 Actually it's the same code with only minor changes in the
+ *	 way the data is stored; this is to support the abstraction
+ *	 of an optional secure memory allocation which may be used
+ *	 to avoid revealing of sensitive data due to paging etc.
+ *	 The GNU MP Library itself is published under the LGPL;
+ *	 however I decided to publish this code under the plain GPL.
+ */
+
+#include "mpi-internal.h"
+#include "longlong.h"
+
+#ifndef UMUL_TIME
+#define UMUL_TIME 1
+#endif
+#ifndef UDIV_TIME
+#define UDIV_TIME UMUL_TIME
+#endif
+
+
+mpi_limb_t
+mpihelp_mod_1(mpi_ptr_t dividend_ptr, mpi_size_t dividend_size,
+			mpi_limb_t divisor_limb)
+{
+	mpi_size_t i;
+	mpi_limb_t n1, n0, r;
+	mpi_limb_t dummy __maybe_unused;
+
+	/* Botch: Should this be handled at all?  Rely on callers?	*/
+	if (!dividend_size)
+		return 0;
+
+	/* If multiplication is much faster than division, and the
+	 * dividend is large, pre-invert the divisor, and use
+	 * only multiplications in the inner loop.
+	 *
+	 * This test should be read:
+	 *	 Does it ever help to use udiv_qrnnd_preinv?
+	 *	   && Does what we save compensate for the inversion overhead?
+	 */
+	if (UDIV_TIME > (2 * UMUL_TIME + 6)
+			&& (UDIV_TIME - (2 * UMUL_TIME + 6)) * dividend_size > UDIV_TIME) {
+		int normalization_steps;
+
+		normalization_steps = count_leading_zeros(divisor_limb);
+		if (normalization_steps) {
+			mpi_limb_t divisor_limb_inverted;
+
+			divisor_limb <<= normalization_steps;
+
+			/* Compute (2**2N - 2**N * DIVISOR_LIMB) / DIVISOR_LIMB.  The
+			 * result is a (N+1)-bit approximation to 1/DIVISOR_LIMB, with the
+			 * most significant bit (with weight 2**N) implicit.
+			 *
+			 * Special case for DIVISOR_LIMB == 100...000.
+			 */
+			if (!(divisor_limb << 1))
+				divisor_limb_inverted = ~(mpi_limb_t)0;
+			else
+				udiv_qrnnd(divisor_limb_inverted, dummy,
+						-divisor_limb, 0, divisor_limb);
+
+			n1 = dividend_ptr[dividend_size - 1];
+			r = n1 >> (BITS_PER_MPI_LIMB - normalization_steps);
+
+			/* Possible optimization:
+			 * if (r == 0
+			 * && divisor_limb > ((n1 << normalization_steps)
+			 *		       | (dividend_ptr[dividend_size - 2] >> ...)))
+			 * ...one division less...
+			 */
+			for (i = dividend_size - 2; i >= 0; i--) {
+				n0 = dividend_ptr[i];
+				UDIV_QRNND_PREINV(dummy, r, r,
+						((n1 << normalization_steps)
+						 | (n0 >> (BITS_PER_MPI_LIMB - normalization_steps))),
+						divisor_limb, divisor_limb_inverted);
+				n1 = n0;
+			}
+			UDIV_QRNND_PREINV(dummy, r, r,
+					n1 << normalization_steps,
+					divisor_limb, divisor_limb_inverted);
+			return r >> normalization_steps;
+		} else {
+			mpi_limb_t divisor_limb_inverted;
+
+			/* Compute (2**2N - 2**N * DIVISOR_LIMB) / DIVISOR_LIMB.  The
+			 * result is a (N+1)-bit approximation to 1/DIVISOR_LIMB, with the
+			 * most significant bit (with weight 2**N) implicit.
+			 *
+			 * Special case for DIVISOR_LIMB == 100...000.
+			 */
+			if (!(divisor_limb << 1))
+				divisor_limb_inverted = ~(mpi_limb_t)0;
+			else
+				udiv_qrnnd(divisor_limb_inverted, dummy,
+						-divisor_limb, 0, divisor_limb);
+
+			i = dividend_size - 1;
+			r = dividend_ptr[i];
+
+			if (r >= divisor_limb)
+				r = 0;
+			else
+				i--;
+
+			for ( ; i >= 0; i--) {
+				n0 = dividend_ptr[i];
+				UDIV_QRNND_PREINV(dummy, r, r,
+						n0, divisor_limb, divisor_limb_inverted);
+			}
+			return r;
+		}
+	} else {
+		if (UDIV_NEEDS_NORMALIZATION) {
+			int normalization_steps;
+
+			normalization_steps = count_leading_zeros(divisor_limb);
+			if (normalization_steps) {
+				divisor_limb <<= normalization_steps;
+
+				n1 = dividend_ptr[dividend_size - 1];
+				r = n1 >> (BITS_PER_MPI_LIMB - normalization_steps);
+
+				/* Possible optimization:
+				 * if (r == 0
+				 * && divisor_limb > ((n1 << normalization_steps)
+				 *		   | (dividend_ptr[dividend_size - 2] >> ...)))
+				 * ...one division less...
+				 */
+				for (i = dividend_size - 2; i >= 0; i--) {
+					n0 = dividend_ptr[i];
+					udiv_qrnnd(dummy, r, r,
+						((n1 << normalization_steps)
+						 | (n0 >> (BITS_PER_MPI_LIMB - normalization_steps))),
+						divisor_limb);
+					n1 = n0;
+				}
+				udiv_qrnnd(dummy, r, r,
+						n1 << normalization_steps,
+						divisor_limb);
+				return r >> normalization_steps;
+			}
+		}
+		/* No normalization needed, either because udiv_qrnnd doesn't require
+		 * it, or because DIVISOR_LIMB is already normalized.
+		 */
+		i = dividend_size - 1;
+		r = dividend_ptr[i];
+
+		if (r >= divisor_limb)
+			r = 0;
+		else
+			i--;
+
+		for (; i >= 0; i--) {
+			n0 = dividend_ptr[i];
+			udiv_qrnnd(dummy, r, r, n0, divisor_limb);
+		}
+		return r;
+	}
+}
+
+/* Divide num (NP/NSIZE) by den (DP/DSIZE) and write
+ * the NSIZE-DSIZE least significant quotient limbs at QP
+ * and the DSIZE long remainder at NP.	If QEXTRA_LIMBS is
+ * non-zero, generate that many fraction bits and append them after the
+ * other quotient limbs.
+ * Return the most significant limb of the quotient, this is always 0 or 1.
+ *
+ * Preconditions:
+ * 0. NSIZE >= DSIZE.
+ * 1. The most significant bit of the divisor must be set.
+ * 2. QP must either not overlap with the input operands at all, or
+ *    QP + DSIZE >= NP must hold true.	(This means that it's
+ *    possible to put the quotient in the high part of NUM, right after the
+ *    remainder in NUM.
+ * 3. NSIZE >= DSIZE, even if QEXTRA_LIMBS is non-zero.
+ */
+
+mpi_limb_t
+mpihelp_divrem(mpi_ptr_t qp, mpi_size_t qextra_limbs,
+	       mpi_ptr_t np, mpi_size_t nsize, mpi_ptr_t dp, mpi_size_t dsize)
+{
+	mpi_limb_t most_significant_q_limb = 0;
+
+	switch (dsize) {
+	case 0:
+		/* We are asked to divide by zero, so go ahead and do it!  (To make
+		   the compiler not remove this statement, return the value.)  */
+		/*
+		 * existing clients of this function have been modified
+		 * not to call it with dsize == 0, so this should not happen
+		 */
+		return 1 / dsize;
+
+	case 1:
+		{
+			mpi_size_t i;
+			mpi_limb_t n1;
+			mpi_limb_t d;
+
+			d = dp[0];
+			n1 = np[nsize - 1];
+
+			if (n1 >= d) {
+				n1 -= d;
+				most_significant_q_limb = 1;
+			}
+
+			qp += qextra_limbs;
+			for (i = nsize - 2; i >= 0; i--)
+				udiv_qrnnd(qp[i], n1, n1, np[i], d);
+			qp -= qextra_limbs;
+
+			for (i = qextra_limbs - 1; i >= 0; i--)
+				udiv_qrnnd(qp[i], n1, n1, 0, d);
+
+			np[0] = n1;
+		}
+		break;
+
+	case 2:
+		{
+			mpi_size_t i;
+			mpi_limb_t n1, n0, n2;
+			mpi_limb_t d1, d0;
+
+			np += nsize - 2;
+			d1 = dp[1];
+			d0 = dp[0];
+			n1 = np[1];
+			n0 = np[0];
+
+			if (n1 >= d1 && (n1 > d1 || n0 >= d0)) {
+				sub_ddmmss(n1, n0, n1, n0, d1, d0);
+				most_significant_q_limb = 1;
+			}
+
+			for (i = qextra_limbs + nsize - 2 - 1; i >= 0; i--) {
+				mpi_limb_t q;
+				mpi_limb_t r;
+
+				if (i >= qextra_limbs)
+					np--;
+				else
+					np[0] = 0;
+
+				if (n1 == d1) {
+					/* Q should be either 111..111 or 111..110.  Need special
+					 * treatment of this rare case as normal division would
+					 * give overflow.  */
+					q = ~(mpi_limb_t) 0;
+
+					r = n0 + d1;
+					if (r < d1) {	/* Carry in the addition? */
+						add_ssaaaa(n1, n0, r - d0,
+							   np[0], 0, d0);
+						qp[i] = q;
+						continue;
+					}
+					n1 = d0 - (d0 != 0 ? 1 : 0);
+					n0 = -d0;
+				} else {
+					udiv_qrnnd(q, r, n1, n0, d1);
+					umul_ppmm(n1, n0, d0, q);
+				}
+
+				n2 = np[0];
+q_test:
+				if (n1 > r || (n1 == r && n0 > n2)) {
+					/* The estimated Q was too large.  */
+					q--;
+					sub_ddmmss(n1, n0, n1, n0, 0, d0);
+					r += d1;
+					if (r >= d1)	/* If not carry, test Q again.  */
+						goto q_test;
+				}
+
+				qp[i] = q;
+				sub_ddmmss(n1, n0, r, n2, n1, n0);
+			}
+			np[1] = n1;
+			np[0] = n0;
+		}
+		break;
+
+	default:
+		{
+			mpi_size_t i;
+			mpi_limb_t dX, d1, n0;
+
+			np += nsize - dsize;
+			dX = dp[dsize - 1];
+			d1 = dp[dsize - 2];
+			n0 = np[dsize - 1];
+
+			if (n0 >= dX) {
+				if (n0 > dX
+				    || mpihelp_cmp(np, dp, dsize - 1) >= 0) {
+					mpihelp_sub_n(np, np, dp, dsize);
+					n0 = np[dsize - 1];
+					most_significant_q_limb = 1;
+				}
+			}
+
+			for (i = qextra_limbs + nsize - dsize - 1; i >= 0; i--) {
+				mpi_limb_t q;
+				mpi_limb_t n1, n2;
+				mpi_limb_t cy_limb;
+
+				if (i >= qextra_limbs) {
+					np--;
+					n2 = np[dsize];
+				} else {
+					n2 = np[dsize - 1];
+					MPN_COPY_DECR(np + 1, np, dsize - 1);
+					np[0] = 0;
+				}
+
+				if (n0 == dX) {
+					/* This might over-estimate q, but it's probably not worth
+					 * the extra code here to find out.  */
+					q = ~(mpi_limb_t) 0;
+				} else {
+					mpi_limb_t r;
+
+					udiv_qrnnd(q, r, n0, np[dsize - 1], dX);
+					umul_ppmm(n1, n0, d1, q);
+
+					while (n1 > r
+					       || (n1 == r
+						   && n0 > np[dsize - 2])) {
+						q--;
+						r += dX;
+						if (r < dX)	/* I.e. "carry in previous addition?" */
+							break;
+						n1 -= n0 < d1;
+						n0 -= d1;
+					}
+				}
+
+				/* Possible optimization: We already have (q * n0) and (1 * n1)
+				 * after the calculation of q.  Taking advantage of that, we
+				 * could make this loop make two iterations less.  */
+				cy_limb = mpihelp_submul_1(np, dp, dsize, q);
+
+				if (n2 != cy_limb) {
+					mpihelp_add_n(np, np, dp, dsize);
+					q--;
+				}
+
+				qp[i] = q;
+				n0 = np[dsize - 1];
+			}
+		}
+	}
+
+	return most_significant_q_limb;
+}
+
+/****************
+ * Divide (DIVIDEND_PTR,,DIVIDEND_SIZE) by DIVISOR_LIMB.
+ * Write DIVIDEND_SIZE limbs of quotient at QUOT_PTR.
+ * Return the single-limb remainder.
+ * There are no constraints on the value of the divisor.
+ *
+ * QUOT_PTR and DIVIDEND_PTR might point to the same limb.
+ */
+
+mpi_limb_t
+mpihelp_divmod_1(mpi_ptr_t quot_ptr,
+		mpi_ptr_t dividend_ptr, mpi_size_t dividend_size,
+		mpi_limb_t divisor_limb)
+{
+	mpi_size_t i;
+	mpi_limb_t n1, n0, r;
+	mpi_limb_t dummy __maybe_unused;
+
+	if (!dividend_size)
+		return 0;
+
+	/* If multiplication is much faster than division, and the
+	 * dividend is large, pre-invert the divisor, and use
+	 * only multiplications in the inner loop.
+	 *
+	 * This test should be read:
+	 * Does it ever help to use udiv_qrnnd_preinv?
+	 * && Does what we save compensate for the inversion overhead?
+	 */
+	if (UDIV_TIME > (2 * UMUL_TIME + 6)
+			&& (UDIV_TIME - (2 * UMUL_TIME + 6)) * dividend_size > UDIV_TIME) {
+		int normalization_steps;
+
+		normalization_steps = count_leading_zeros(divisor_limb);
+		if (normalization_steps) {
+			mpi_limb_t divisor_limb_inverted;
+
+			divisor_limb <<= normalization_steps;
+
+			/* Compute (2**2N - 2**N * DIVISOR_LIMB) / DIVISOR_LIMB.  The
+			 * result is a (N+1)-bit approximation to 1/DIVISOR_LIMB, with the
+			 * most significant bit (with weight 2**N) implicit.
+			 */
+			/* Special case for DIVISOR_LIMB == 100...000.  */
+			if (!(divisor_limb << 1))
+				divisor_limb_inverted = ~(mpi_limb_t)0;
+			else
+				udiv_qrnnd(divisor_limb_inverted, dummy,
+						-divisor_limb, 0, divisor_limb);
+
+			n1 = dividend_ptr[dividend_size - 1];
+			r = n1 >> (BITS_PER_MPI_LIMB - normalization_steps);
+
+			/* Possible optimization:
+			 * if (r == 0
+			 * && divisor_limb > ((n1 << normalization_steps)
+			 *		       | (dividend_ptr[dividend_size - 2] >> ...)))
+			 * ...one division less...
+			 */
+			for (i = dividend_size - 2; i >= 0; i--) {
+				n0 = dividend_ptr[i];
+				UDIV_QRNND_PREINV(quot_ptr[i + 1], r, r,
+						((n1 << normalization_steps)
+						 | (n0 >> (BITS_PER_MPI_LIMB - normalization_steps))),
+						divisor_limb, divisor_limb_inverted);
+				n1 = n0;
+			}
+			UDIV_QRNND_PREINV(quot_ptr[0], r, r,
+					n1 << normalization_steps,
+					divisor_limb, divisor_limb_inverted);
+			return r >> normalization_steps;
+		} else {
+			mpi_limb_t divisor_limb_inverted;
+
+			/* Compute (2**2N - 2**N * DIVISOR_LIMB) / DIVISOR_LIMB.  The
+			 * result is a (N+1)-bit approximation to 1/DIVISOR_LIMB, with the
+			 * most significant bit (with weight 2**N) implicit.
+			 */
+			/* Special case for DIVISOR_LIMB == 100...000.  */
+			if (!(divisor_limb << 1))
+				divisor_limb_inverted = ~(mpi_limb_t) 0;
+			else
+				udiv_qrnnd(divisor_limb_inverted, dummy,
+						-divisor_limb, 0, divisor_limb);
+
+			i = dividend_size - 1;
+			r = dividend_ptr[i];
+
+			if (r >= divisor_limb)
+				r = 0;
+			else
+				quot_ptr[i--] = 0;
+
+			for ( ; i >= 0; i--) {
+				n0 = dividend_ptr[i];
+				UDIV_QRNND_PREINV(quot_ptr[i], r, r,
+						n0, divisor_limb, divisor_limb_inverted);
+			}
+			return r;
+		}
+	} else {
+		if (UDIV_NEEDS_NORMALIZATION) {
+			int normalization_steps;
+
+			normalization_steps = count_leading_zeros(divisor_limb);
+			if (normalization_steps) {
+				divisor_limb <<= normalization_steps;
+
+				n1 = dividend_ptr[dividend_size - 1];
+				r = n1 >> (BITS_PER_MPI_LIMB - normalization_steps);
+
+				/* Possible optimization:
+				 * if (r == 0
+				 * && divisor_limb > ((n1 << normalization_steps)
+				 *		   | (dividend_ptr[dividend_size - 2] >> ...)))
+				 * ...one division less...
+				 */
+				for (i = dividend_size - 2; i >= 0; i--) {
+					n0 = dividend_ptr[i];
+					udiv_qrnnd(quot_ptr[i + 1], r, r,
+						((n1 << normalization_steps)
+						 | (n0 >> (BITS_PER_MPI_LIMB - normalization_steps))),
+						divisor_limb);
+					n1 = n0;
+				}
+				udiv_qrnnd(quot_ptr[0], r, r,
+						n1 << normalization_steps,
+						divisor_limb);
+				return r >> normalization_steps;
+			}
+		}
+		/* No normalization needed, either because udiv_qrnnd doesn't require
+		 * it, or because DIVISOR_LIMB is already normalized.
+		 */
+		i = dividend_size - 1;
+		r = dividend_ptr[i];
+
+		if (r >= divisor_limb)
+			r = 0;
+		else
+			quot_ptr[i--] = 0;
+
+		for (; i >= 0; i--) {
+			n0 = dividend_ptr[i];
+			udiv_qrnnd(quot_ptr[i], r, r, n0, divisor_limb);
+		}
+		return r;
+	}
+}